The 2010 Table - Magic Numbers, Tragic Numbers & Other Numerical Arcana of Dubious Value [R]

It’s that time of year again …

Code:

[B][U]         GP  GR W   D   L   PTS PPG   PRJ MPP SE# M#  T#  HD#[/U]
01  LAG  24  6  14  05  05  47  1.96  59  65  19  05  33  -35
02  CMB  24  6  13  05  06  44  1.83  55  62  16  08  30  -20
03  RSL  24  6  12  08  04  44  1.83  55  62  16  08  30  -20
04  FCD  23  7  10  11  02  41  1.78  53  62  16  11  30  -05
05  NYR  24  6  12  04  08  40  1.67  50  58  12  12  26  00
06  SJE  22  8  09  06  07  33  1.50  45  57  11  19  25  27
07  COL  23  7  09  07  07  34  1.48  44  55  09  18  23  26
08  SEA  24  6  09  06  09  33  1.38  41  51  05  19  19  35
-------------------------------------------------------------
09  KCW  22  8  07  06  09  27  1.23  37  51  05  25  19  49
10  CHI  22  8  06  08  08  26  1.18  35  50  04  26  18  53
11  TFC  24  6  07  07  10  28  1.17  35  46  00  24  14  60
12  CDC  23  7  07  04  12  25  1.09  33  46  00  27  14  65
13  NER  23  7  07  03  13  24  1.04  31  45  -01 28  13  69
14  PHI  23  7  06  06  11  24  1.04  31  45  -01 28  13  69
15  HOU  23  7  06  05  12  23  1.00  30  44  -02 29  12  73
16  DCU  24  6  05  03  16  18  0.75  23  36  -10 34  04  110

Code:

GP = Games Played | GR = Games Remaining | PPG = Points Per Game
MPP = Max Points Possible = PTS + (3 x GR)
SE# = SS Elimination Number = MPP - (1st Place PTS) + 1 
M# = Magic Number = (9th Highest MPP) – PTS + 1
T# = Tragic Number = MPP – (8th Highest PTS) + 1
HD# = Historical Despair Number = Full explanation below

- Sorted by PPG, then fewest GP, then alphabetical.*
- Teams in [COLOR="SlateGray"]gray[/COLOR] have been eliminated from the playoffs.
- Teams in [COLOR="DarkBlue"]blue[/COLOR] have clinched a playoff spot.
- Teams in [I]italics[/I] are conference champions.

* In the final weeks MLS tie-breakers will replace the final alphabetical 
sort. But until the final weeks it's seriously not worth the bother.[/B]

Code:

[B][U]Pace: Deviation from 3 at home, 1 on the road[/U]

[U]         GP  PTS HG  AG  HW HD HL AW AD AL PCE[/U]
01  LAG  24  47  11  13  7  2  2  7  3  3  01
02  FCD  23  41  11  12  7  3  1  3  8  1  -04
03  CMB  24  44  12  12  9  1  2  4  4  4  -04
04  RSL  24  44  12  12  9  3  0  3  5  4  -04
05  NYR  24  40  12  12  8  1  3  4  3  5  -08
07  SJE  22  33  11  11  5  3  3  4  3  4  -11
06  COL  23  34  11  12  6  4  1  3  3  6  -11
08  KCW  22  27  11  11  5  3  3  2  3  6  -17
09  SEA  24  33  13  11  6  3  4  3  3  5  -17
10  CHI  22  26  11  11  3  6  2  3  2  6  -18
11  CDC  23  25  11  12  5  2  4  2  2  8  -20
12  PHI  23  24  11  12  4  5  2  2  1  9  -21
13  TFC  24  28  13  11  6  5  2  1  2  8  -22
14  NER  23  24  12  11  6  2  4  1  1  9  -23
15  HOU  23  23  12  11  5  3  4  1  2  8  -24
16  DCU  24  18  12  12  3  1  8  2  2  8  -30

Sorted by PCE, then fewest GP, then PTS.[/B]

Code:

[B][U]PPG: Home and Away[/U]

[U]         HPPG             APPG[/U]
01  RSL  2.50    01  LAG  1.85
02  CMB  2.33    02  FCD  1.42
03  FCD  2.18    03  SJE  1.36
04  LAG  2.09    04  CMB  1.33
05  NYR  2.08    05  NYR  1.25
06  COL  2.00    06  RSL  1.17
07  TFC  1.77    07  SEA  1.09
08  NER  1.67    08  COL  1.00
09  SJE  1.64    09  CHI  1.00
10  KCW  1.64    10  KCW  0.82
11  SEA  1.62    11  CDC  0.67
12  CDC  1.55    12  DCU  0.67
13  PHI  1.55    13  PHI  0.58
14  HOU  1.50    14  TFC  0.45
15  CHI  1.36    15  HOU  0.45
16  DCU  0.83    16  NER  0.36[/B]

Code:

[B][U]Remaining League Schedule[/U]

[U]      WK25  WK26  WK27  WK28  WK29  WK30[/U]
CDC:  KCW   @PHI  LAG   TFC   @SEA  SJE
      ---   ---   ---   ---   ---   CHI
CHI:  @RSL  SEA   @SJE  CMB   DCU   @CDC
      ---   ---   @FCD  ---   ---   ---
CMB:  SEA   @NER  SJE   @CHI  @TFC  PHI
COL:  NER   @RSL  DCU   @FCD  @LAG  RSL
DCU:  @LAG  HOU   @COL  SJE   @CHI  TFC
FCD:  NYR   NER   CHI   COL   @RSL  @LAG
      ---   @KCW  ---   ---   ---   ---
HOU:  TFC   @DCU  @PHI  NER   @SJE  SEA
KCW:  @CDC  FCD   @NYR  SEA   @NER  SJE
LAG:  DCU   NYR   @CDC  @PHI  COL   FCD
NER:  @COL  @FCD  RSL   @HOU  KCW   @NYR
      ---   CMB   ---   ---   ---   ---
NYR:  @FCD  @LAG  KCW   RSL   @PHI  NER
PHI:  @SJE  CDC   HOU   LAG   NYR   @CMB
RSL:  CHI   COL   @NER  @NYR  FCD   @COL
SEA:  @CMB  @CHI  TFC   @KCW  CDC   @HOU
SJE:  PHI   @TFC  CHI   @DCU  HOU   CDC
      ---   ---   @CMB  ---   ---   @KCW
TFC:  @HOU  SJE   @SEA  @CDC  CMB   @DCU[/B]

Notes on The Non-Existent Playoff Race (And the Despair Numbers Explained)

For the last few years we’ve been blessed with some great MLS playoff races. I hope you all enjoyed them because this year’s playoff race is shaping up to be a real dud. Why do I say this? Let me apologize for my lengthy explanation in advance …

For most of the season over on the DC United board I’ve been tracking something that I call the Despair Numbers, which are an estimate of the pace a team would require for the remainder of their season to end the season as a serious playoff contender. There are two types.

The first (and most important) type is the Historical Despair Number (HD#), and it’s calculated like this:

HD# = [(40 points - Actual Points) / Games Remaining] * 30​

A number of people have noticed that for the last several years the lower minimum for playoff viability has been about 40 points. That’s not to say you’re in with 40 points, only that 40 points should put you within historical spitting distance of the playoffs. The HD# is based on this observation. The HD# means: a team must play like a “##” point team for the remainder of the season to meet the 40 point threshold.

(Important Note On Rounding: For the Despair Numbers it really only makes sense to round up because if you round down you're actually aiming below the bar. For example, if you need 1.38 PPG to hit 40 points, then you're actually aiming for a 41.4 point pace. Round down to 41 and you won't hit 40. You have to aim for a 42 point pace to clear the bar.)

People will quibble about this 40 point threshold. That’s OK. It’s anecdotal and should be treated as such. But to those who might doubt the validity or utility of the HD# I’d ask one question: Do you honestly believe any teams below 40 or so points will have any shot whatsoever to make the playoffs this year? Yeah, that’s what I thought …

// skip this unless you’re really bored ... or want to be //

The historical 40 point threshold is something of a lower-minimum for playoff viability. But it does not reflect the actual performance of the mid-table teams, which although unlikely to fall short of a 40 point pace might actually exceed a 40 point pace. That’s why I’ve also been tracking a second Despair Number that I’ve been calling the Actual Despair Number (AD#). It’s calculated like this:

AD# = [([(Current 8th Place Points Per Game * 30) + 1] - Actual Points) / Games Remaining] * 30​

This number reflects the actual performance of the mid-table teams. Rank the teams by points per game, project the 8th place team's points per game over a 30 game season, add 1 (because you need to beat that team), subtract the team’s actual points, divide by the games remaining, multiply by 30. The AD# means: a team must play like a “##” point team for the remainder of the season to outpace and thus beat the current 8th place team's actual pace.

You can tell some interesting stories with both of these numbers, but to my mind the HD# is the more compelling number for two reasons. First, while the actual pace of the mid-table teams has at times climbed up to the mid-40s (and briefly into the high-40s), it’s actually tended to hover in the low 40s and as I write this it’s exceeding the historical 40 point threshold by only a single point. Second, as a practical matter the AD# functions as something of an upper minimum for playoff viability, but there’s really no point in worrying about the upper minimum if your team can’t even meet the lower minimum. That’s why I think the HD# tells the critical story, while the AD# really only provides a bit of color. I intend to treat it as such, posting it and discussing it only if it actually adds something (which, at the moment, it doesn’t).

// now back to the main deal //

Let’s look at the table. Four teams have already surpassed 40 points and have negative HD#s. A fifth team stands exactly at 40 points and therefore has an HD# of zero. All five of these teams are either already virtually in the playoffs (LAG, CMB, RSL and FCD) or they’re already virtually standing in the threshold of playoff viability (NYR).

Then you have SJE and COL, teams that are currently on pace to end the season at 45 and 44 points respectively. But the HD#s say two things. First, there’s an enormous gap between NYR and the SJE and COL pair (which shows you just how secure the top five teams really are). Second, both SJE and COL can severely under-perform in comparison to their current pace and still squeak into playoff viability. SJE is playing like a 45 point team, but to hit 40 points they really only need to finish out the season at a 27 point pace. COL is playing like a 44 point team, but to hit 40 points they really only need to finish out the season at a 26 point pace. In other words, both of these teams could have near meltdowns and still (probably) stumble into the playoffs.

Turning to the bottom of the table DCU’s HD# is 110. Because the highest possible HD# in a 30 game season is 90, any HD# above 90 means it is impossible for the team to meet the 40 point threshold. Playoff-wise, DCU is dead and buried.

Above DCU are TFC, CDC, NER, PHI and HOU. Technically it remains possible for these teams to hit 40 points. But the paces required range from virtually impossible to totally absurd. TFC tops the bunch and would have to complete their season at a 60 point pace to hit 40 points. Not even the best team in the league is playing at a 60 point pace. At the other end of this bunch, just above DCU, is HOU with the totally absurd HD# of 73. Barring some sort of crazy 2009 FCD type run (and recall that FCD ended that season at 39 points and failed to qualify for the playoffs), all five of these teams are essentially zombies - the walking dead.

So we’ve got LAG, CMB, RSL, FCD and NYR at the top of the table. Then we’ve got SJE and COL comfortably below them. That’s 7 playoff spots already. At the other end of the table there’s DCU, HOU, PHI, NER, CDC and TFC who are all already effectively eliminated from playoff contention. That’s 6 teams already gone. Add all them together and we’ve accounted for 7 playoff spots and 13 teams.

The remaining teams are SEA, KCW and CHI. In terms of a playoff race, these three teams are really the only contestants, and they’re really only vying for a single spot. All of them have HD#s that are in the realm of reasonable possibility. SEA is obviously in the best position because they’re the only one of the three with a tiny bit of slack: their HD# remains less than their actually projected current pace. But KCW and CHI, with HD#s of 49 and 53 respectively, are both in positions where a couple of good games would transform them into legitimate contenders. At the same time SEA is positioned to pull out of reach with a good result or two, and KCW and CHI have no room for error.

Sure, sometimes the unexpected happens. It’s still possible that the 2010 playoff race will take some strange turn and become a really exciting race to the finish. But that’s not at all what I expect to happen. Barring something crazy (fingers crossed!), it’s likely to wrap up in the next couple of weeks. And that’d really stink.

Housekeeping & Other Rubbish

Because the playoff race is likely to suck, in the long run I’d like to broaden the focus of this thread in comparison to its previous incarnations so that it is less about the playoffs specifically and more about the table and the various numbers in general.

Errors: They happen. Alert me to them in this thread. (Seriously, double check me on the schedules.)

Full updates once a week, mini-updates (the main table) as warranted.

Future posts in this thread won’t be anywhere near this long!

The DC United Dungeon of Doom & Doggies!

Carrying a couple things over from this thread, here are a few words on DC United’s current situation. With its victory over TFC, United has escaped the indignity of vying for the title of the Worst MLS Team EVER! United required 17 points to best the ‘01 Mutiny once and for all, and today the team stands at 18 points. As shown here, the ’01 TB# (the pace needed to beat out the 2001 Mutiny) has dropped below zero. More, as you can see here, the Olsen Pace is once again above the Onalfo Line.

That said, besting the ’00 United (the team’s previous worst season) would require 11 points from the remaining 6 games (about a 51 point pace), and that’s just not very likely. Still, the game against TFC was arguably the biggest win of the season since it saved the team from a fate that a number of us have been worrying about for several months. As such, today we’ve got a rare happy puppy!

 

I think you've got a completely legitimate point -- and, yes, the despair numbers really are just riffs on PPG style pace numbers. And when I started thinking about those numbers in relation to DCU (just this year, by the way), I initially thought of them in terms of PPG. The reason I decided (essentially) to multiply those PPG numbers by 30 was for two reasons.

1) Perhaps you're an exception, but I don't think most people (like me, for instance) really "get" the implications of needing, let's say, 2.00 PPG. What do I mean by that? People don't usually think in PPG, but they do know what a 60 PT team is like. You say my team needs to complete the season at a 2.00 PPG pace to be viable for the playoffs and I'm going to think that sounds sort of possible, even approachable. But if you say my team needs to complete the season at a 60 PT pace, then I'm gonna start looking for a drink because I know that's virtually out of the realm of realistic possibility.

But those numbers are the same. It's just that one number captures the actual magnitude of the situation, while the other seems to diminish it. Perhaps this is because most people seem more easily to grasp the full implications of larger numbers and analogies (you must play like a 60 point team), but they have trouble sensing the real implications of raw numbers, especially small ones with decimal places. (The despair number also has the advantage accentuating the large scale implications of what might appear to be minor differences of a few decimal points in PPG.)

2) Related to the first, because multiplying these numbers by 30 emphasizes and accentuates the magnitude of the numbers (both alone and in comparison with other teams) it's much, much easier to tell something of a story with them. To this extent it was a stylistic choice in favor of the number that offered a better narrable quality.

All that said, I understand exactly what you're saying and where you're coming from.

---

Adding one thing -- I do think as the season really closes the despair numbers will lose their narrable quality. But at that point the whole measure might be moot anyway.

 

Yes. But he's right (like I noted above), when we really are just talking about short runs, the measure will lose its force.

 

For the sake of comparison ...

Code:

[B]         GP  GR  W   D   L   PTS HD#  HD# in PPG
01  LAG  24  06  14  05  05  47  -35  -1.17
02  CMB  24  06  13  05  06  44  -20  -0.67
03  RSL  24  06  12  08  04  44  -20  -0.67
04  FCD  23  07  10  11  02  41  -05  -0.17
05  NYR  24  06  12  04  08  40  00   0.00
07  COL  23  07  09  07  07  34  26   0.87
06  SJE  22  08  09  06  07  33  27   0.90
08  SEA  24  06  09  06  09  33  35   1.17
09  KCW  22  08  07  06  09  27  49   1.63
10  CHI  22  08  06  08  08  26  53   1.77
11  TFC  24  06  07  07  10  28  60   2.00
12  CDC  23  07  07  04  12  25  65   2.17
13  NER  23  07  07  03  13  24  69   2.30
14  PHI  23  07  06  06  11  24  69   2.30
15  HOU  23  07  06  05  12  23  73   2.43
16  DCU  24  06  05  03  16  18  110  3.67[/B]

On controversies like this, I'm usually tempted just to include both numbers. I'll see how it looks on the next update.

Personally, I'm still not a fan of the PPG measure. It just doesn't speak to me the same way. Plus, I think I'd rather err toward over-accentuating the magnitudes than understating them.

 

Shoot! I'm actually missing several games in that schedule, and not just the ones identified above. I think/hope this is now correct, but no guarantees.

Code:

[B][U]Remaining League Schedule[/U]

[U]      WK25  WK26  WK27  WK28  WK29  WK30[/U]
CDC:   KCW  @PHI   LAG   TFC  @SEA   SJE
      ----  ----  ----  ----  ----   CHI
CHI:  @RSL   SEA  @SJE   CMB   KCW  @CDC
      ----  ----  @FCD  ----   DCU  ----
CMB:   SEA  @NER   SJE  @CHI  @TFC   PHI
COL:   NER  @RSL   PHI  @FCD  @LAG   RSL
      ----  ----   DCU  ----  ----  ----
DCU:  @LAG   HOU  @COL   SJE  @CHI   TFC
FCD:   NYR   NER   CHI   COL  @RSL  @LAG
      ----  @KCW  ----  ----  ----  ----
HOU:   TFC  @KCW  @PHI   NER  @SJE   SEA
      ----  @DCU  ----  ----  ----  ----
KCW:  @CDC   HOU  @NYR   SEA   CHI   SJE
      ----   FCD  ----  ----  @NER  ----
LAG:   DCU   NYR  @CDC  @PHI   COL   FCD
NER:  @COL  @FCD   RSL  @HOU   KCW  @NYR
      ----   CMB  ----  ----  ----  ----
NYR:  @FCD  @LAG   KCW   RSL  @PHI   NER
PHI:  @SJE   CDC  @COL   LAG   NYR  @CMB
      ----  ----   HOU  ----  ----  ----
RSL:   CHI   COL  @NER  @NYR   FCD  @COL
SEA:  @CMB  @CHI   TFC  @KCW   CDC  @HOU
SJE:   PHI  @TFC   CHI  @DCU   HOU   CDC
      ----  ----  @CMB  ----  ----  @KCW
TFC:  @HOU   SJE  @SEA  @CDC   CMB  @DCU[/B]

Bad idea to take a shortcut when I initially put this schedule together ...

Like I said, I'll look into adding an HD#/30 column in the next update. But I don't want that chart to get too wide or else it's going to screw up the formatting on some screens. I'll figure it out.

Before moving on, there are actually two other reasons I like the HD# as a 30 game pace. First, and maybe I should have said this, it identifies a 40 point pace as something akin to "par" for playoff viability. If you're on a 40 point pace, then your HD# will be 40 as well. That's a little easier for me to grasp than saying 1.333 PPG is par for playoff viability. Second, I actually developed the number to use in graphs like this one of DCU's unfortunate season. I don't think that chart would come across as well in PPG.

But like you said, people see things in different ways and that's fine. I already include a whole table that to my eye doesn't seem terribly pertinent to MLS (the pace table) because evidently it does seem pretty pertinent to some others. So adding a column isn't a big deal to me as long as the table remains clean and legible.

...

Anyway ... I wanted to highlight some things beyond the playoff race.

1) Here's the final tables from last year. Check out the HPPG numbers. RSL topped the chart in 2009 with 2.13 PPG at home. This year they top the HPPG chart again and they're outpacing their performance last year with a whopping 2.50 PPG. Unreal! Now take a look at their APPG in 2009. They were tied for second to the bottom at .53 PPG. This year they've more than doubled their away pace: it now stands at 1.17 PPG. Sure, that's decent but still pretty modest away record. But the improvement from last year is truly remarkable. To put a positive spin on all of this you might say that if you're not an RSL fan it's pretty scary to think about how good they might be next year if they can maintain this sort of trajectory. A negative spin? RSL's great year is built virtually entirely on its home performance.

2) LA's dominating year, however, is built largely on its road performance. Yes, they are a very good home team, but their away performance leaves everyone else in the league in the dust -- bettering the second best away team in the league by about 30%.

3) This disparity between the top and the bottom teams on the table is much, much greater this year than it has been in recent years -- and especially compared to last year. There's a bunch of very good teams crowded around the top of the table and a bunch of truly bad teams converging at the bottom. It's not just the points and PPG disparity that strikes me here, it's the fact that with about a quarter of the season to go, the entire bottom third of the table has been eliminated from SS contention. If 2010 is any indication, then parity is dead.